基于PNN和IGS的铅锌烧结块成分智能集成预测模型

An intelligent integrated-prediction model for components of Pb-Zn agglomerate based on the process neural network(PNN) and the improved grey system(IGS)

中文关键词  铅锌烧结过程  成分预测  过程神经网络  改进灰色系统  信息熵  智能集成预测模型

英文关键词  lead-zinc sintering process  prediction of component  process neural network  improved grey system  information entropy  intelligent integrated-prediction model

中文摘要

      针对复杂的烧结块成分预测问题, 提出一种基于过程神经网络和改进灰色系统的铅锌烧结块成分智能集成预测模型. 首先利用过程神经网络可充分表达时间序列中时间累积效应、灰色系统可弱化数据序列波动性的特点, 分别对烧结块成分进行预测, 然后从信息论的观点出发, 提出一种确定各预测模型加权系数的熵值递推算法, 通过对两个预测模型的预测结果进行加权集成, 获得更加准确的铅锌烧结块成分预测结果. 结果表明, 智能集成模型 的预测精度高于单一预测模型, 能有效地对烧结块成分进行预测, 满足了配料计算对预测精度和数据完备性的

英文摘要

      To deal with the problem of the component prediction for Pb-Zn agglomerate, an intelligent integratedprediction model based on the process neural network(PNN) and the improved grey system(IGS) is presented. First, the component of agglomerate is predicted by PNN and IGS models, and then, a recursive entropy algorithm for the weighting coefficients is devised from the viewpoint of the information theory. The component of Pb-Zn agglomerate is predicted by integrating the two prediction models. Application results show that the integrated model has high prediction accuracy; it predicts the components of agglomerate efficiently and meets the data-completeness requirements for proportioning computation.